Outline of the Earth Simulator Project

Transcription

1 Message from the Director The Earth Simulator has contributed greatly to the prediction of global warming and the interpretation of various climate change phenomena. At the same time, through these activities, the Earth Simulator has played an important role in the area of the promotion of the research and human resources development which is imperative for the sustainable progress of the science and technology. CEIST/JAMSTEC is determined to create a new terrain of Marine-Earth Informatics based on the accumulation of the vast technological and operational knowledge thus to challenge to combine the data and information of Marine-Earth Science with the innovation. Dr. Keiko Takahashi Director-General of Center for Earth Information Science and Technology (CEIST) Outline of the Earth Simulator Project 1. Mission and Basic Principles of the Earth Simulator The Earth Simulator was developed for the following aims. The first aim is to ensure a bright future for human begins by accurately predicting variable global environment. The second is to contribute to the development of science and technology in the 21st century. Based on these aims, the principles listed below are established for the projects of the Earth Simulator. 1) Each project should be open to researches in each research field and to the public, rather than it is confined within the limited research society. 2) In principle, the research achievements obtained by using the Earth Simulator should be promptly published and returned to the public. 3) Earth project should be carried out for peaceful purposes only. 2. Earth Simulator Research Project The allocation of the Earth Simulator resources for each research field in FY2014 was decided to be as shown in Fig. 1. There are three project categories in the Earth Simulator resource allocation. The projects are shown in Table 1 and 2. 1) Proposed Research projects: We accept and select applications for the research projects of the earth science field which is included Climate Change, Solid Earth, Natural Disasters, etc. 2) Contract Research projects: The projects using the Earth Simulator focus on research commissioned by public organizations such as the government. - The Program for Risk Information on Climate Change (SOUSEI). - The Program for Creating Innovation by Sharing Advanced Research Facilities (The Strategic Industrial Use) - Consignment Study (JST/CREST, KAKENHI, etc.) 3) JAMSTEC Research projects: The Earth Simulator is also used for research projects organized by JAMSTEC, international and domestic collaboration projects and the execution of urgent jobs in the time of natural disasters. In addition, fee-based usage of the Earth Simulator is included in this category. Fig. 1 The Allocation of Resources of the Earth Simulator in FY2014 1

2 Earth Science (21 projects) Table 1 Proposed Research Projects in FY2014 Project Name Name of Project Professional Affiliation of Project 1 Understanding Roles of Oceanic Fine Structures in Climate and its Variability Hideharu Sasaki APL, JAMSTEC 2 Simulations of Adaptation-Oriented Strategy for Climate Variability Ryo Ohnishi CEIST, JAMSTEC 3 Development Research of a High-quality Climate Model for Long-term Climate Change Projection Study Michio Kawamiya ICCP, JAMSTEC 4 Simulations of Atmospheric General Circulations of Earth-like Planets by AFES Yoshiyuki Hayashi Graduate School of Science, Kobe University 5 Study on the Diagnostics and Projection of Marine Ecosystem Change Associated with Global Change Sanae Chiba RCGC, JAMSTEC 6 Study of Cloud and Precipitation Processes using a Global Cloud-system Resolving Model Tomoe Nasuno DSEP, JAMSTEC 7 Study on the Predictability of Climate Variations and Their Mechanisms Swadhin Behera APL, JAMSTEC 8 Improved Ocean State Estimation by using a 4D-VAR Ocean Data Assimilation System Shuhei Masuda RCGC, JAMSTEC 9 Global Elastic Response Simulation Seiji Tsuboi CEIST, JAMSTEC 10 Simulation Study on the Dynamics of the Mantle and Core in Earth-like Conditions Yozo Hamano DEEP, JAMSTEC 11 Numerical Simulation of Seismic Wave Propagation and Strong Ground Motions in 3-D Heterogeneous Media Takashi Furumura Center for Integrated Disaster Information Research, Interfaculty Initiative in Information Studies, The University of Tokyo/Earthquake Research Institute, The University of Tokyo 12 Development of Advanced Simulation Tools for Solid Earth Sciences Mikito Furuichi MAT, JAMSTEC 13 Numerical Simulations of the Dynamics of Volcanic Phenomena Takehiro Koyaguchi Earthquake Research Institute, University of Tokyo 14 Space and Earth System Modeling Kanya Kusano OELE, JAMSTEC 15 Numerical Experiments with Multi-models for Paleoenvironmental Problems Ayako Abe Atmosphere and Ocean Research Institute, The University of Tokyo 16 Predictability Variation in Numerical Weather Prediction Takeshi Enomoto Disaster Prevention Research Institute, Kyoto University 17 Computational Science of Turbulence in Atmospheric Boundary Layers Takashi Ishihara Center for Computational Science, Graduate School of Engineering, Nagoya University 2

3 Project Name Name of Project Professional Affiliation of Project 18 A Large-scale Self-Organizing Map for Metagenome Studies for Surveillance of Microbial Community Structures Toshimichi Ikemura Nagahama Institute of Bio-Science and Technology 19 Generation Mechanism of the Banded Structures Observed in the Jovian-type Planetary Atmospheres Shinichi Takehiro Research Institute for Mathematical Sciences, Kyoto University 20 Analysis of Global Ecosystem Ecology by Fragment Molecular Orbital (FMO) Method Tadashi Maruyama RCMB, JAMSTEC 21 Study for Seamless Prediction of Weather and Climate using Atmosphere-ocean Coupled Global Cloudsystem Resolving Model Kazuyoshi Ohuchi DSEP, JAMSTEC JAMSTEC: Japan Agency for Marine-Earth Science and Technology APL: Application Laboratory CEIST: Center for Earth Information Science and Technology ICCP: Department of Integrated Climate Change Projection Research RCGC: Research and Development (R&D) Center for Global Change DSEP: Department of Seamless Environmental Prediction Research DEEP: Department of Deep Earth Structure and Dynamics Research MAT: Department of Mathematical Science and Advanced Technology OELE: Laboratory of Ocean-Earth Life Evolution Research RCMB: Research and Development (R&D) Center for Marine Biosciences Table 2 Collaboration Projects in FY2014 Collaborative Research on the Sophistication of the Computational Simulation Software toward Constructing the Platform for the Leading Industrial Research and Development: Institute of Industrial Science, The University of Tokyo Ultra High Resolution Simulation for the Safety of International Transportation on the Sea: Kobe University Forecasting thunder cloud: JAXA Numerical simulations of scalar transfer across wind-driven air-water interface: Kyoto University Rapid Change of the Arctic Climate System and its Global Influences: NIPR Implementation Agreement between ESC/JAMSTEC and DOPS/IFREMER on simulation research using the ES: IFREMER JAMSTEC-IPRC Collaborative Study on Climate Change and variability in the Asia-Pacific region : IPRC 3

4 3. System Configuration of the Earth Simulator The Earth Simulator (ES2) is the upgrade system of the previous Earth Simulator, which has significantly Table 3 Real Application Benchmarks Performance contributed to the development of a simulation culture in the area of earth science and related technical fields and introduces new features to bring accurate and high-speed analysis and projections of global-scale environmental phenomena. The current ES is also used to product numerical simulations for advanced research fields that are beyond the scope of other computing systems. By the examination at the time of procurement, the average performance of real application benchmarks exceeded the first system s twice (Table 3). ES2 is in the independent network environment and can be used via the Internet under advanced security protection (Fig. 2). In the Earth Simulator network, there are ES2, Login Servers, User Disks, Terminals and other servers. The current ES2 is a highly parallel vector supercomputer system of the distributed-memory type, and consisted of 160 processor nodes connected by Fat-Tree Network. Each Processor nodes is a system with a shared memory, consisting of 8 vector-type arithmetic processors, a 128GB main memory system. The peak performance of each Arithmetic processor is 102.4Gflops. The ES as a whole thus consists of 1280 arithmetic processors with 20 TB of main memory and the theoretical performance of 131Tflops. All of the software available on the ES2 system are designed and developed so that users can fully and readily exploit the outstanding performance of the world's Fig. 2 ES System Outline largest vector- type computer. ES2 is basically a batch-job system. Network Queuing System II (NQSII) is introduced to manage the batch job. The L batch queue is majored for a production run. The nodes allocated to a L batch queue are used exclusively for that batch job to estimate the job termination time and to make it easy to allocate nodes for the next batch jobs in advance. The batch job is scheduled based on elapsed time instead of CPU time to an efficiently job execution. The job can use the nodes exclusively and the processes in each node can be executed simultaneously. As a result, the large-scale parallel program is able to be executed efficiently. Earth Simulator 2 (March 2009 February 2015) 4

5 4. Next Generation The platform system of Earth Simulator was replaced in March 2015 for the second time following the first replacement in March The simulation performance of the new system in earth science is about 10 times higher than the former one. The new system can process simulations that handle complicated parameters, which have been difficult with the former one, and run larger scale simulations faster. It is expected to contribute solving global environment issues, elucidating the mechanism of crustal movement and earthquake mechanism, and predicting tsunami damage. New Earth Simulator (March 2015 ) 5

6